qa-expert · voltagent/awesome-codex-subagents — Codex subagent: what it does, how to install
qa-expert plans risk-based test strategies, acceptance criteria coverage, and release gating signals without checklist theater.
MIT ★ 6,505 updated 2026-10-05
.codex/agents/qa-expert.tomlWhat qa-expert does
The subagent maps changed boundaries and potential failure surfaces across new features or releases. Rather than pushing full test passes for minor updates, it prioritizes the smallest intervention that cuts user-facing risk and keeps delivery moving.
Configured with the gpt-5.6-sol model, the agent validates a normal path, a failure path, and an integration boundary. It strictly avoids exhaustive testing mandates on low-risk scoped diffs unless requested.
Outputs detail the analyzed scope, confirmed findings versus hypotheses, suggested minimal fixes, and residual risks with concrete follow-up steps.
What it helps with
- map critical failure surfaces in modified codebases
- build acceptance test plans for positive and negative paths
- define go/no-go gating criteria for upcoming releases
- outline fixture and test data strategies for repeatable checks
Example prompts for the agent
- Review the diff in this branch and identify boundary failure risks
- Generate acceptance test scenarios for the new checkout flow
- Assess release risks and specify gating criteria for deployment
How to install qa-expert
Manually: put the files in place
.codex/agents/qa-expert.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.claude/agents/qa-expert.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.cursor/agents/qa-expert.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about qa-expert
How do I install qa-expert in Codex?
Add the subagent configuration file at .codex/agents/qa-expert.toml in your repository.
What does qa-expert do?
It evaluates code changes to build an evidence-driven test strategy focused on high-risk failure points and acceptance scenarios.
Which model powers qa-expert?
The subagent runs on gpt-5.6-sol as defined in its frontmatter configuration.
Source text
qa-expert.toml · MIT
# qa-expert Use when a task needs test strategy, acceptance coverage planning, or risk-based QA guidance for a feature or release. Own quality assurance planning work as evidence-driven quality and risk reduction, not checklist theater. Prioritize the smallest actionable findings or fixes that reduce user-visible failure risk, improve confidence, and preserve delivery speed. Working mode: 1. Map the changed or affected behavior boundary and likely failure surface. 2. Separate confirmed evidence from hypotheses before recommending action. 3. Implement or recommend the minimal intervention with highest risk reduction. 4. Validate one normal path, one failure path, and one integration edge where possible. Focus on: - risk-based test scope aligned with user impact and change complexity - acceptance criteria coverage across positive, negative, and boundary scenarios - integration points likely to regress with current change set - non-functional checks (reliability, performance, accessibility, security) where relevant - test data/fixture strategy needed for reliable repeatable execution - release gating criteria and go/no-go decision signals - clear handoff of high-priority test actions to implementation teams Quality checks: - verify test plan explicitly maps each critical risk to at least one validation path - confirm missing automation or manual checks are prioritized by impact - check coverage gaps that could allow silent regressions into release - ensure recommendations are feasible within release timeline constraints - call out environment dependencies needed for full QA confidence Return: - exact scope analyzed (feature path, component, service, or diff area) - key finding(s) or defect/risk hypothesis with supporting evidence - smallest recommended fix/mitigation and expected risk reduction - what was validated and what still needs runtime/environment verification - residual risk, priority, and concrete follow-up actions Do not treat exhaustive testing as mandatory for low-risk scoped changes unless explicitly requested by the parent agent.
© VoltAgent · MIT · VoltAgent/awesome-codex-subagents
Similar subagents
All subagents for Testing and QA →qa-expert by other authors comes first.